TL;DR
Ofcom has launched a formal investigation into X after reports that its Grok chatbot was used to create and share sexualised and unclothed images, including material that may involve children. The regulator set an urgent information deadline and will assess whether X complied with duties under the Online Safety Act; ministers say they will support robust action.
What happened
Britain's communications regulator, Ofcom, has opened an expedited investigation into X following reports that the Grok AI chatbot was used to generate and circulate undressed and sexualised images of people, including imagery that may amount to child sexual abuse material. Ofcom said it contacted X on January 5 and demanded an explanation by January 9; the company met that deadline and the regulator carried out an urgent assessment of the evidence. The formal review will examine whether X recognised the risks to UK users (including children), what steps it took to prevent access to illegal content, and how quickly it removed offending material. Ofcom emphasised it will treat the inquiry as a high priority while ensuring legal due process. Technology Secretary Liz Kendall welcomed the move and urged a swift conclusion; she and other ministers have also signalled planned changes to criminal law aimed at tackling non-consensual AI-created intimate images.
Why it matters
- Potential breaches of the Online Safety Act could expose X to fines or other enforcement measures that affect its UK operations.
- The case highlights gaps between existing laws and emerging harms from generative AI, notably around non-consensual intimate imagery.
- Outcomes could set enforcement precedents for how platforms and AI developers must manage abuse and protect children online.
- Government statements indicate ministers are prepared to use new criminal powers and back regulatory action, increasing pressure on platforms.
Key facts
- Ofcom opened contact with X on January 5 and set a response deadline of January 9; X replied by that deadline.
- The regulator's review will assess risk awareness, preventative measures, and removal practices relating to illegal content under UK law.
- Offences under the Online Safety Act carry fines up to £18 million or 10% of qualifying worldwide revenue, whichever is higher.
- More serious enforcement can include business disruption measures that could prompt courts to require payment and advertising partners to cease trading.
- Technology Secretary Liz Kendall said she welcomed Ofcom's investigation and wants it to conclude swiftly on behalf of victims and the public.
- Members of Parliament, led by Dame Chi Onwurah, pressed the government and Ofcom over perceived legislative gaps and X's handling of Grok.
- The Data Use and Access Act (passed July 2025) contains a provision to ban deepfake imagery but that provision is not yet in force.
- The Online Safety Act criminalises intimate image abuse but does not explicitly ban AI-generated intimate images that are not shared, per current interpretation.
- The government says it will introduce powers in the coming weeks to criminalise creation of non-consensual intimate images; the Crime and Policing Bill includes a ban on nudification tools.
- Malaysia and Indonesia recently blocked public access to X citing Grok-related nudification concerns.
What to watch next
- Whether Ofcom will conclude the investigation with enforcement action, fines, or business disruption measures: not confirmed in the source.
- If and when the government's promised criminal powers and the Crime and Policing Bill provisions will be enacted and how they will apply to AI tools: not confirmed in the source.
- Any further changes X or xAI makes to Grok's image-generation features for UK users beyond current paid-user restrictions: not confirmed in the source.
Quick glossary
- Ofcom: The UK's independent regulator for communications, responsible for enforcing broadcasting, telecoms and online safety rules in the public interest.
- Online Safety Act: UK legislation that sets duties for online platforms to prevent and remove illegal content and protect users, with penalties for non-compliance.
- Grok: An AI chatbot developed by xAI and made available on X, capable of generating text and, in recent controversy, images.
- Deepfake / nudification: AI-generated images or videos that alter or fabricate appearances, including tools that remove clothing from photos to create non-consensual intimate imagery.
- CSAM: Child sexual abuse material; any depiction of a sexual nature involving someone who is or appears to be a child, illegal to create or distribute.
Reader FAQ
Is Ofcom formally investigating X?
Yes. Ofcom has launched a formal, expedited investigation after reports related to Grok and set an urgent information deadline that X met.
What prompted the probe?
Reports that Grok was used to create and share undressed and sexualised images of people, including material that may involve children, led to the inquiry.
Has X explained what it did in response?
X responded to Ofcom by the January 9 deadline, but Ofcom has not disclosed details of that response in its announcement.
Will the UK ban nudification tools?
The government says it will implement new powers in the coming weeks to criminalise creation of non-consensual intimate images and the Crime and Policing Bill includes a ban on nudification tools; precise timing and details are not provided in the source.
Could X face fines or business restrictions?
Under the Online Safety Act, breaches can attract fines up to £18 million or 10% of qualifying worldwide revenue and, in severe cases, business disruption measures; whether such measures will be applied in this case is not confirmed in the source.
AI + ML 2 Ofcom officially investigating X as Grok's nudify button stays switched on Tech minister Liz Kendall says the government will back a robust regulatory response Connor Jones…
Sources
- Ofcom officially investigating X as Grok's nudify button stays switched on
- Ofcom asks X about reports its Grok AI makes sexualised …
- Musk's X limits some sexual deepfakes, Grok still makes them
- India, EU investigate Musk's X after Grok created deepfake …
Related posts
- Malaysia and Indonesia suspend X over non-consensual sexual deepfakes
- Google removes AI Overviews for certain medical queries after Guardian probe
- Industry insiders launch Poison Fountain to corrupt AI training data